ABSTRACT
Support vector machines (SVMs) are one of the most preferred algorithms for performing classification tasks. Their aim is to find a hyperplane among the data points belonging to different classes with the aid of support vectors. If datasets cannot be classified linearly, then the kernel trick is used instead to map the features to higher dimensions, where they can subsequently be classifiable. In this chapter, we will describe the implementation of SVMs in Python using the “sklearn” package and in performing a grid search for choosing optimized parameters. This was accomplished for predicting two classification tasks - one for the classification of the heart disease dataset; and the other for the prediction of wheat species based on the seed dataset. The Jupyter Notebook file with a template for the execution of classification tasks using a support vector machine is shared as a supplementary file with this chapter.
